股市风险VaR与ES的动态度量与分析

首先描述金融时间序列的一般特性，从收益率的波动性与分布两方面进行考虑，建立起计算时变风险值VaR和ES的模型，并在多种分布情形下动态测算上证综合指数的风险，结果表明基于GED分布的VaR模型能够较好地刻画高频时间序列的尖峰肥尾性及杠杆效应等特性，而ES模型则有效地弥补了VaR模型的不足之处。     

有偏胖尾分布下的金融市场风险测度方法

通过对上证综指和世界股市若干重要指数收益的统计特征分析发现,无论是成熟资本市场还是新兴资本市场,其收益分布都展现出较为显著的“有偏”和“尖峰胖尾”特征,因此,主流金融理论假定的正态分布或对称学生分布都无法全面准确刻画股市收益的真实分布特征和风险状况。通过引入有偏的学生分布,分析和对比了不同收益分布假定下的市场波动率和风险价值计算方法,并利用风险价值的失败率似然比检验以及动态分位数回归检验法,实证分析了不同分布模型的适用范围和精确程度,探讨了非正态分布假定下的金融市场风险测度方法。     

Two ARCH Models and Their Limitations as Diffusion Processes

IntroductionMany financial time series such as stock returnsand exchange rates exhibit an important feature,conditional heteroskedasticity,which means thatmarket volatilities tend to cluster.This featurewas described in many financial fields as early asClark[1] ,Merton[2 ] and others.Engle[3 ] firstintroduced the ARCH model which is now widelyused.An autoregressive model of order p with anARCH error term of order q can be written asYt=μ +1Yt-1+… +p Yt-p +εt (1 )εt|Ψt-1～ N (0 ,…     

基于偏态分布的风险度量计算

金融时间序列具有尖峰厚尾性,同时在股市中又存在着杠杆效应.对股票指数收盘价格的对数收益率序列建立ARMA-APARCH模型,在对数收益率序列分别满足Skewed-t分布和Skewed-GED的假设下,给出了在险价值及期望损失的计算方法.对t分布与Skewed-t分布、GED与Skewed-GED分别进行对比性实证分析,结果表明,在两个偏态分布假设下计算得到的期望损失估计结果更为保守,更能够捕捉到股市的尾部风险.     

证券组合SKST-APARCH模型和VaR估计分析

金融资产无条件收益分布表现为偏度和峰度共存，运用ARCH模型族在波动和Valt分析中往往将对称的后尾t、GED误差分布与不对称冲击结合在一起来解决序列不对称特性．由于偏度和峰度并非相互独立．这样未能根本上解决问题。本文在APARCH模型中引入偏t分布（Skst）模型对上证指数和深证A股指数的波动及VaR进行了实证分析，并与Gauss、t、GED等对称分布下的估计进行了比较，结果显示Skst能较好的拟舍波动、较准估计VaR．     

Estimating and Forecasting APARCH‐Skew‐t Model by Wavelet Support Vector Machines

This paper concentrates on comparing estimation and forecasting ability of quasi‐maximum likelihood (QML) and support vector machines (SVM) for financial data. The financial series are fitted into a family of asymmetric power ARCH (APARCH) models. As the skewness and kurtosis are common characteristics of the financial series, a skew‐t distributed innovation is assumed to model the fat tail and asymmetry. Prior research indicates that the QML estimator for the APARCH model is inefficient when the data distribution shows departure from normality, so the current paper utilizes the semi‐parametric‐based SVM method and shows that it is more efficient than the QML under the skewed Student's‐t distributed error. As the SVM is a kernel‐based technique, we further investigate its performance by applying separately a Gaussian kernel and a wavelet kernel. The results suggest that the SVM‐based method generally performs better than QML for both in‐sample and out‐of‐sample data. The outcomes also highlight the fact that the wavelet kernel outperforms the Gaussian kernel with lower forecasting error, better generation capability and more computation efficiency. Copyright © 2014 John Wiley & Sons, Ltd.